## wp17247

## Source details

**Canonical URL:** [wp17247](https://www.imf.org/-/media/files/publications/wp/2017/wp17247.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2017/wp17247.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2017/wp17247.pdf.json)

---

### Introduction and context
- Governments face conflicting objectives in taxing the cellular telecommunications (“telecom”) industry: promote wide, cheap, and rapidly modernizing services versus capture significant revenue from large and growing turnovers.
- Telecom services are important inputs into productivity and growth because of possible externality and social inclusion effects and their role in delivering services such as banking, health, and education.

### Key empirical and fiscal observations
- Telecom companies are major taxpayers in many low- and middle-income countries:
  - Jamaica: total telecoms revenues (including license and spectrum fees) averaged 1.8 percent of GDP in 2013-2015.
  - Malawi: total telecoms revenues (including license and spectrum fees) averaged 1.3 percent of GDP in 2013-2015.
  - Senegal: mobile phone companies accounted for 20 to 30 percent of corporate income tax between 2005 and 2009.
  - Haiti: over a quarter of the country’s sales tax (excluding customs) came from the telecoms sector in 2014.
- Aggregate taxation trends:
  - The tax burden on mobile telecommunications increased from 17.4 percent in 2007 to 20.1 percent in 2015 (Deloitte), with the sample size rising from 101 countries in 2007 to 110 countries in 2015.
  - In 2016, developed countries had 3.3 times as many cell phone subscriptions as fixed-lined subscriptions, while developing countries had 11.3 times as many.

### Composition of the total tax burden on telecoms
- Total tax burden includes:
  - consumer, corporate, and trade taxes;
  - sector-specific taxes such as CIT surcharges, VAT surcharges, service and handset excises;
  - elevated customs charges on capital equipment;
  - regulatory charges, notably spectrum and/or operator license fees.
- Sample sector-specific provisions (country examples preserved as in source):
  - CIT surcharge: Cote d'Ivoire, Jamaica, Jordan, Tunisia, Yemen, Zambia.
  - VAT surcharge: Argentina, Brazil, Jamaica, Mauritania, Sudan.
  - Excises (domestic calls): Albania, Bangladesh, Dominican Republic, Ecuador, Greece, Jamaica, Kenya, Malawi, Nepal, Pakistan, Tanzania, Turkey, Uganda, Ukraine, Venezuela, Zambia.
  - Excises (international calls): Congo-Brazzaville, Gabon, Ghana, Jamaica, Malawi, Senegal.
  - Handsets: Bangladesh, Botswana, Brazil, Ghana, Nigeria, Lesotho, Syria, Tunisia, Turkey.
  - Connection fees: India, Tanzania.
  - Tax incentives/reduced tax rates: Angola, Congo-Brazzaville, Niger.
- Sources for tax provision inventory include GSMA (2012), Deloitte (2016a, 2016b), dutycalculator.com.

### Sector-specific taxation: rationale and concerns
- Prevalence and analogies:
  - Sector-specific taxes in telecoms have become as prevalent as in extractive industries.
  - Extractive-sector higher taxes are justified by economic rents from inelastically supplied, non-renewable resources; the paper investigates whether similar conditions hold for telecommunications.
- Concerns highlighted:
  - Industry claims of over-taxation and potential negative growth effects from rising tax burdens.
  - Developing countries may rely more on easy-to-administer taxes (e.g., excises) given lower administrative capacity, possibly leading to heavier sector-specific taxation than in developed countries.

### II. Economics of the telecommunications sector — Network Externalities
- Conceptual points:
  - Telecommunications is a network industry: new users generate positive externalities by broadening the network (Economides, 1996).
  - Positive externalities imply free markets could yield sub-optimal cell phone penetration; subsidies (e.g., for handsets) or cross-subsidies for remote areas can be efficient.
  - Common policy responses: cross-subsidizing network expansion, buildout requirements, universal service funds (noting governance problems and risk of under-/mis-utilization).
- Empirical evidence (examples preserved):
  - Doganoglu and Grzybowski (2007): "700 percent expansion of cell phone communications over 1998-2003" in Germany largely due to network effects versus "41 percent price decrease."
  - Grajek (2010) finds intra-operator network effects for Poland (1996-2001).
  - Diffusion-model studies (Wu and Chu (2009), Rouvinen (2006)) find both network effects (existing penetration) and potential user base significantly impact diffusion.
- Penetration statistics:
  - Ratio of global phone subscriptions (fixed and mobile) to population: "107 percent in developing countries and 165 percent in developed countries."
  - Unique mobile phone subscribers as a percentage of global population: "67 percent worldwide—62 percent in developing countries."
- Policy implication: public policy likely to continue supporting network expansion to underserved areas; universal service funds financed by telecom charges can be inefficient.

### II.B Contribution to Economic Growth
- Mechanisms:
  - Cell phones lower information acquisition costs, reduce transaction costs, create opportunities for additional transactions, and improve economic efficiency and growth.
  - Tax measures that slow cell phone penetration can yield short-term revenue but harm long-term growth benefits.
- Macroevidence (exact figures preserved):
  - Waverman, Meschi and Fuss (2005): among low income countries, "10 more mobile phones per 100 people increased per capita GDP growth by 0.6 percentage points—roughly twice as much as in high income countries."
  - Sridhar and Sridhar (2007): "an increase of 1 percent in mobile penetration increased GDP by 0.01 percent"; average cumulative growth in cell phone use of "248 percent increased GDP by 2.48 percent."
  - Lee, Levendis and Gutierrez (2012): significant impact in 2000-2006 for 44 sub-Saharan countries; stronger where landline penetration is lowest.
  - Andrianaivo and Kpodar (2012): "a 10-percentage point increase in the mobile penetration rate increases the growth rate by 0.6 percentage points." They also find the price of a "3-minute conversation is negatively related to growth."
- Microeconomic evidence and case studies (selected figures):
  - Acker (2010): introduction of cell phones in Niger between 2001 and 2006 lowered price dispersion by "10 to 16 percent" on grain markets.
  - Acker and Fafchamps (2015): "6 percent" reduction in producer price dispersion in Niger following cell phone introduction.
- Policy nuance:
  - Evidence generally suggests a positive causal relationship between cell phone use and growth but does not automatically justify government intervention; careful tax design is required to avoid deterring access.

### II.C Demand Elasticity
- Role in tax rationale: excises sometimes justified by inelastic demand offering a good "tax handle."
- Empirical elasticity evidence (diverse estimates preserved exactly):
  - Commerce Commission (2003) review of 13 papers:
    - Network access: "-0.06 to -0.54"
    - Mobile originated calls: "-0.09 to -0.8"
  - Waverman, Meschi and Fuss (2005): "-1.5"
  - Garbacz and Thompson (2007): connection charges: "-0.37 to -0.029"; monthly price: "-1.268 to -0.195"
  - Dewenter and Haucap (2008) Austria traffic:
    - Short-term, businesses: "-0.33"
    - Short-term, private consumers: "-0.14"
    - Long term, businesses: "-0.74"
    - Long term, private consumers: "-0.37"
  - Gasmi et al. (2009) South Africa: "Between -3.8 and -1.3"
  - Caves (2011) USA: "-2.1"
  - Koutroumpis et al. (2011) Greece: "-1.6"
  - Hakim and Neaime (2014) MENA: "Between –1.2 and -1.0"
  - Karacuka et al. (2011) Turkey:
    - Short term, pre-paid: "-0.20"
    - Short term, post-paid: "-0.36"
    - Long term, pre-paid: "-0.33"
    - Long term, post-paid: "-0.72"
- Conclusion: elasticity estimates are diverse; network access elasticity tends to be lower than usage elasticities. Whether inelasticity justifies excise taxation is discussed in tax policy sections.

### II.D Technology, industry structure, and market power
- Historical and structural points:
  - Fixed-wire telephony historically a "natural monopoly"; liberalization since the 1980s transformed the sector toward contestability with cellular networks and VoIP.
  - Entry costs remain substantial; spectrum scarcity constrains provider numbers.
  - Spectrum frequencies noted: "800-900 megahertz (MHz) range" (best suited) and up to "2+ gigahertz (GHz)" sometimes used.
  - Typical market structure: "Most countries have 2-4 providers, while some retain a monopoly structure"; "Less than one fifth of countries have at least 5 providers."
  - Li and Lyons (2012) identify "5 providers and more" as threshold for full effective competition.
- Evidence of rents and profitability (exact figures preserved):
  - US telecom sector profitability since 2009: average return on assets of large US telecom service companies exceeded that of large-capitalization companies in general by "70 percent"; return on investment and return on equity were "16 percent and 15 percent higher than the market, respectively."
  - Boniecki and others (2016) for Middle East and Africa: dominant operators' EBITDA-to-revenue ratios "41-47 percent", second-largest "30-37 percent", third and fourth "15-25 percent".
- Technological and organizational changes (VoIP, data-based communications, virtual mobile network operators) can shift rents and competition depending on regulatory and licensing policies.

### II.E Regulation
- Regulatory policy affects rents via spectrum licensing, license fees, and price regulation.
- Spectrum and market depth:
  - Spectrum policy determines number of mobile telecom licenses; country size positively influences number of operators and efficiency (Symeou, 2011).
  - Market concentration data: average market share of second providers is less than "one third"; third and fourth providers less than "20 percent" and "10 percent", respectively.
- Regulatory interventions:
  - Price regulation may be needed where incumbents control key infrastructure.
  - Buildout requirements and universal service funds extend coverage but reduce operator profitability; universal service funds risk inefficient deployment.
  - Spectrum and license charges often substantial and materially affect profitability.
- Governance considerations:
  - Regulatory charges should finance regulatory costs; charges above cost recovery are taxes.
  - Fiscal authority should be concentrated in the treasury; residual revenue from spectrum licenses or other fees raised by regulators should flow to the treasury to avoid divergence of interest.

### III. Tax policy — General considerations
- Core tax objectives: finance public goods, ensure efficiency, equity, and simplicity — few taxes with low rates on wide, uniform bases.
- Justifications for differential treatment:
  - Industries generating economic rents can be efficiently taxed without distorting investment incentives.
  - Presence of externalities may justify subsidies or taxes.
  - Administrative infeasibility of standard taxes may justify second-best instruments.
- Telecom-specific points:
  - Technological and regulatory barriers to entry, limited spectrum, and small market sizes can generate rents—suggesting some suitability for rent taxation.
  - Profit-based taxes (CIT, rent taxes) may be difficult due to profit-shifting by MNEs; low-capacity countries may prefer simpler instruments like excises despite distortion.
  - Positive externalities argue for avoiding taxes that specifically limit access.
- Public finance principle: taxes on productive inputs should be avoided; customs duties on capital equipment should ideally be zero.

### III.B Spectrum license fees
- Form and duration:
  - Spectrum license fees are usually fixed payments levied up-front and/or periodically; typical license duration "about 15 years."
- Allocation methods and effects:
  - Competitive auction preferred since 1990s; auctions allocate licenses to efficient providers and procure revenue.
  - Theoretical expectation: bidders pay up to the rents from holding the license.
- Empirical findings and risks (preserved language):
  - Some studies (Morris (2005), Kwerel (2000), Bauer (2003)) find no evidence license fees impact service pricing.
  - More recent work indicates auctions may affect pricing and penetration: Janssen and Karamychev (2009); Haan and Toolsema (2011); Kuroda and del Pilar Baquero Forero (2017) find penetration rates are lower in countries that conduct auctions.
- Risks and design issues:
  - Maximizing auction revenue can perversely limit competition, raise consumer prices, and encourage collusion.
  - Faulty auction design, excessive reserve values, and collusion can cause auction failure.
- Valuation complexity and alternatives:
  - License valuation depends on spectrum amount, market size, license terms, and transfer rights.
  - Ex-ante fixed fees shift risk to providers and may cause shaded bids.
  - Alternatives: ex-post profit- or gross-revenue sharing (royalties) distribute risk and can create incentives to expand allocation and investment.
  - Kwon, Lee and Oh (2010) favor royalties over ex-ante fixed fees.
- Policy recommendation: improve government capacity for accurate spectrum valuation; consider auction design, reserve values, and risk-sharing fee structures.

### III.C Rent Taxes (including Box 1: Rent Taxation Options)
- Concept and illustrative example:
  - Resource rent taxes (RRTs) tax profits above the “normal rate of return.” Example: if normal rate of return is "15 percent", an equity-financed investment of US$100 yielding profit US$25 generates a rent of US$10; CIT base is US$25, RRT base is US$10.
- Application to telecoms:
  - Authors note RRTs have not yet been applied to the telecom industry (to their knowledge).
- Rent-based tax designs summarized (names and key features preserved):
  - Brown Tax or R-based cash flow tax: state as passive investor; immediate expensing; cash-flow basis; Australian RSPT proposal approximated this tax (Henry, and others 2010).
  - Resource rent tax: annual uplift on accumulated losses until recovered; uplift rate set at minimum required rate of return.
  - Variable Income Tax (VIT): varies CIT rate by profits-to-revenue ratio; developed for South African gold mining.
  - Tax surcharge on cash flow: add back depreciation and interest, deduct capital expenditure in full; example: U.K. North Sea petroleum CIT surcharge rate from 2011/12 is 32 percent in addition to normal CIT.
  - Allowance for Corporate Capital (ACC) surcharge scheme: annual uplift on undepreciated capital assets; Norway uses this to tax hydropower rents with a rate of 30 percent added to CIT (CIT rate "24 percent in 2017 and then 23 percent in 2018").
- Efficiency, neutrality, and administrative considerations:
  - RRTs tax only rent and are less distortive than CIT.
  - Well-structured rent taxes should not affect marginal investment, production, or pricing.
  - Rent taxes increase average tax rate on investment if they raise total taxes paid relative to pre-tax profits, potentially affecting entry decisions and spectrum valuation.
  - Rent taxes are more equitable than auctions (based on actual rents) and less vulnerable to collusion than weak auctions.
  - Vulnerabilities: susceptible to accounting manipulation, transfer pricing; administrative costs of introducing a new tax; interaction with existing CIT can interfere with neutrality (Mintz (2017)).
- CIT implications (key findings preserved):
  - Higher CIT rates used in some countries: Jamaica, Côte-d’Ivoire, Jordan and Yemen subject telecom companies to higher CIT rates.
  - CIT taxes the normal return to capital, raising cost of capital and depressing investment.
  - Davidson and Martin (1985) show higher CIT rates may augment oligopolistic collusion.
  - Heckemeyer and Overesch (2013): average semi-elasticity of multinational corporate profits to the CIT rate of -0.8 (a 1 percentage-point higher CIT rate will reduce reported profits of an affiliate by 0.8 percent).
  - Comparative assessment: spectrum auctions, rent taxes, and CIT surcharges all can tax telecom rents; auctions tax expected rents, rent taxes are efficient but defer revenue, CIT surcharges are distortionary but administratively cheaper where CIT systems exist.

### Telecom excises: rationale, effects, and evidence
- Rationale:
  - Excises provide up-front revenue and are easy to administer; function similarly to royalties in extractive industries.
- Economic distortions and externalities:
  - Excises raise prices and suppress consumption; telecom services have positive externalities and are not standard excisable goods.
  - Because excises tax gross revenues, they can render marginal projects unprofitable and reduce investment and output.
- Evidence and country examples (figures preserved):
  - Malawi: domestic excises account for almost half of total fiscal and regulatory revenues generated by telecoms, which averaged 1.3 percent of GDP in 2013-2015.
  - Jamaica: telephone excises plus the VAT surcharge account for more than one third of total telecom revenues, which averaged almost 1.8 percent of GDP in 2013-2015.
  - Jamaican VAT on telecom service is 25 percent, vs. the standard VAT rate of 16.5 percent; the Jamaican VAT surcharge represents 34 percent of VAT revenue, or 0.22 percent of GDP.
- Price incidence and market-structure effects:
  - Incidence depends on supply and demand elasticities; medium-term disinvestment can reduce capacity and raise prices.
  - Excises can accelerate substitution to VoIP; many countries have extended excise base to include internet data (examples: Turkey, Guinea).
  - In limited-competition markets, excises may increase dominant firms' market power by driving out higher-cost operators and facilitating collusion and “overshifting.”
- Guinea case study:
  - Tax introduced: 1 GNF per second (60 GNF per minute, or nearly 15 percent of the average price per minute in Q2-2015).
  - Traffic decreased by 16.1 percent in Q3-2015.
  - Quarterly revenue from the tax is roughly 12 M USD (exchange rate of 10,000 GNF/USD).
  - SMS component had an even more dramatic impact.
- Trade-offs:
  - Moderate royalties/excises can share fiscal and economic risks between government and investors.
  - Excises are less effective than rent taxes or CIT at capturing rents, and they distort consumption and business inputs.

### Specific versus ad valorem excises: welfare and price effects
- Literature conclusion: ad valorem excises dominate specific excises in terms of welfare effects in markets with limited competition.
- Mechanism:
  - Ad valorem excise flattens effective demand curve; specific excise shifts it downward.
  - Ad valorem excise gives producers more incentive to reduce prices because part of a price cut reduces government revenue.
- Practical advantage: Ad valorem excises are straightforward to extend to internet data (megabytes vs. calling minutes).

### Excises on international calls: prevalence, impacts, and distortions
- Trend and examples (figures preserved):
  - Some governments taxed incoming international calls at rates far exceeding domestic call excises.
  - Burundi raised the rate of its termination fee to 247 percent of the underlying charge (GSMA, 2014).
  - Jamaica: excise on domestic calls about US$0.003 per minute, while rate on international calls is US$0.075 per minute.
- Empirical evidence on traffic elasticity:
  - Ghana and Gabon: taxes that raised incoming call prices by 58 and 82 percent led to call volume drops of 27 and 57 percent, respectively (GSMA, 2014; OECD, 2014).
- Technological factors increasing elasticity of the tax base:
  - Shift to internet calling (VoIP) and use of illegal “SIM boxes” that re-route calls.
  - GSMA (2014) estimate: in Ghana, 10 percent of international calls were re-routed through SIM boxes.
- Political and economic costs:
  - Incoming call excises fall heavily on diaspora and business partners; can reduce CIT revenues and remittances.
  - Regional policy response example: May 2014 agreement among Kenya, Rwanda, Burundi, Uganda, and South Sudan not to tax each other’s outgoing calls.
- Policy recommendation: limit international call excises to the same rate as domestic telephone excises to avoid distortions and eliminate incentive to use SIM boxes.

### International call termination and interaction with internet data
- Call routing summary (example preserved): Caller’s handset → local tower → switching point → submarine fiber optic cable (e.g., Africa Coast to Europe, “ACE” cable) → arrival in destination country → transfer back to cell network/land line → end user.
- Tax implementation: taxes on international incoming calls generally imposed via a minimum price to be charged for connecting (“terminating”) the call; tax remitted to destination government as share of that minimum price.
- Interaction with internet data and avoidance: VoIP can bypass incoming-call excises; illegal SIM boxes convert VoIP into local calls subject only to domestic excises.

### Excises on network access (handsets and connection charges)
- Some countries tax up-front costs (handsets, SIM cards).
- Harms:
  - Excises on phone acquisition and initial connection can prevent access—especially in low-income countries—reducing network value.
  - Because of network externalities, initial access should be subsidized rather than taxed.
  - High excises/customs duties on small portable items promote smuggling.
  - Taxes on handsets and initial connection seldom raise significant revenues and should be eliminated.

### Value-added Tax (VAT) on telecoms: design issues and practical adaptations
- VAT application:
  - Most countries apply VAT to telephone services at the standard rate.
  - Input VAT crediting means VAT is not a cost for registered VAT payers; informal companies still experience it as a cost.
  - To raise the same revenue from a VAT surcharge as from an excise, the surcharge must be higher because of input crediting.
- Risks of VAT surcharges and dual rates:
  - Dual VAT rates are difficult to contain and can “creep” to other goods and services.
  - IMF supports a single-rate VAT.
  - Excess VAT credits require refunds, exposing administrative weaknesses.
- Place-of-consumption and international call VAT complexities:
  - Implementing VAT on international calls is complex; misapplication can create double taxation or no taxation.
  - Some countries applied VAT to net international revenue (access fees charged to, minus access fees paid to foreign carriers), effectively applying VAT to net exports; removing such practices has met fierce resistance in low-income countries.
- Practical adaptation example (prepaid cards, Senegal figures preserved exactly):
  - Final consumer price: 1000 CFA Francs (FCFA) with an 18 percent VAT.
  - Pre-VAT price: 1000/1.18 = 847.46 FCFA.
  - VAT amount: 152.54 FCFA.
  - Operator might sell the card for pre-VAT price of 750 FCFA to the bulk distributor, to which the VAT on the final retail price is added, total 902.54 FCFA.
  - Bulk distributor might then resell the card VAT-free to the street vendor for 960 FCFA, who then resells it to the final consumer for 1000 FCFA.
  - Effective VAT rate on the first transaction: 152.54/750, i.e., 20.3 percent.
  - Administrative rationale: collecting VAT at the operator level on the nominal recharge amount avoids administrative difficulties along the distribution chain.

### VAT on prepaid cards and place-of-consumption rules (Box 4 summary)
- Principle: VAT on telephone calls should apply at the point of consumption (destination principle); in practice, determining the point of consumption is complex.
- EU recent revisions recommend treating the country of origin of the calling SIM card as the place of consumption (Council Implementing Regulation (EU) No 1042/2013).
- Implications preserved:
  - Origin-based rule simplifies VAT treatment by deeming calls consumed in the country of the calling SIM card and has implications where VAT-able prepaid/caller-pays systems exist.
  - Origin-based norm implies VAT cannot be applied to net income from Home–Abroad operator transactions as current practice in some countries.
- VAT treatment by scenario (table entries preserved as scenarios):
  - Home / Home / Home / Base case: Originating Home SIM pays VAT; caller’s operator pays a VAT to the receiver’s operator and credits it as input VAT.
  - Home / Home / Abroad / Outgoing: Originating Home SIM pays VAT; Home operator pays access fee to foreign operator and reverse charges the VAT at Home rate; input VAT can be credited.
  - Home / Abroad / Home or Abroad / Outgoing roaming: Service consumed Abroad but deemed consumed at Home because call from a Home SIM; Home VAT applies; Home operator reverse charges VAT at Home rate; input VAT can be credited.
  - Abroad / Home / Home or Abroad / Incoming roaming: Service consumed at Home but deemed consumed Abroad because call from Abroad SIM; Foreign operator pays access fee to Home operator, who exports a service, and the transaction is zero-rated.
  - Abroad / Abroad / Home / Incoming: Home receiver does not pay the VAT.
  - Abroad / Abroad / Abroad / NA: NA.

### Input taxes, user fees, and tax incentives
- Input taxes and user fees:
  - Countries levy charges on telecom inputs (elevated customs duties, user fees for public inputs, charges for infrastructure use, issuance of phone numbers, monitoring minute counts).
  - Efficient to charge for public services when cost can be calculated and exclusive user identified; otherwise fees are input taxes.
  - Input taxes can worsen limited competition and be overshifted to consumers (Katz and Rosen (1985)).
  - Input taxes distort production and should be avoided in favor of efficient taxes; regulatory fees should cover service costs only.
- Tax incentives:
  - Governments often offer telecom operators general tax incentives (e.g., reduced taxes, investment incentives, customs waivers). Examples: Niger (initial 8-year holiday for CIT and other taxes), Central African Republic.
  - Many developing-country incentives are ill-advised, sacrificing scarce revenues for investment that would occur without incentives.
  - Customs waivers on capital goods can be justified on productive-efficiency grounds; preferable to specify zero tariff on capital goods in customs code to minimize discretion.
  - Telecom operator licenses confer exclusive rights to exploit “locational rents”; investment incentives for telecoms are difficult to justify and likely to cause revenue losses.
  - Network externalities may justify lenient terms for establishing networks in new areas; initial cellular licenses were often granted at no charge.
  - Telecommunications is not labor-intensive; incentives aimed at boosting employment are not well-targeted to telecoms.

### Conclusions and policy guidance (pecking-order approach)
- Observations:
  - Taxes on the telecoms sector have proliferated, increasing uncertainty and compliance costs (elevated customs duties, diverse regulatory fees, elevated CIT and VAT rates, telephone call excises).
  - Where rents from limited competition cannot be addressed through regulatory measures, they may justify higher taxation, but many instruments applied do not target rents well and distort markets.
- Recommended “pecking order” of tax instruments (preferences and trade-offs preserved):
  - Auctions: capture rents without distorting investment and prices; favored where regulatory capacity exists. Risks: auctions reflect expected rather than actual rents, collusion/governance issues, impossibility where licenses already awarded, and risk of encouraging monopolistic behavior if revenue-maximization is over-emphasized.
  - Rent tax or CIT surcharge: capture rents with minimal impact on prices; rent taxes are more equitable and flexible than auctions because based on actual profits. Weaknesses: earnings manipulation, transfer pricing; CIT raises cost of capital (reduces investment), rent taxes defer revenues and have higher administrative cost.
  - Taxes on consumption (service excises): imperfect for targeting rents and raise prices/restrict access; simple to design and administer and hard for firms to avoid. If used, ad valorem excises preferable to specific excises because less likely to be over-shifted to consumers. VAT surcharge may be preferable to ad valorem excise since it does not burden formal businesses but may necessitate a higher consumer rate and complicate administration. Excises on handsets and initial connection should be avoided to protect network expansion externalities.
  - Avoid taxes on producer inputs (customs duties, excessive public user fees) as they do not target rents effectively and distort markets.
  - Avoid tax incentives for telecoms as unnecessary to attract investment to exploit the local consumer market; treat telecom taxes same as for other sectors except where clear, limited justifications exist (e.g., buildout requirements, universal service funds).
- Overall guidance:
  - Adopt a balanced tax regime sharing fiscal and economic risks between investors and government.
  - Exercise caution and adaptability given rapid technological change; persistence and allocation of telecom rents will depend on regulatory regimes and market dynamics encouraging effective competition.

*Source: wp17247 (IMF working paper excerpt).*

### References .............................................................................................................

### References

### Introduction and context
- Governments face conflicting objectives in taxing the cellular telecommunications (“telecom”) industry: promote wide, cheap, and rapidly modernizing services versus capture significant revenue from large and growing turnovers.
- Telecom services are considered important inputs into productivity and growth because of possible externality and social inclusion effects, and their role in delivering services such as banking, health, and education.

### Key empirical and fiscal observations
- Telecom companies are major taxpayers in many low- and middle-income countries:
  - Jamaica: total telecoms revenues (including license and spectrum fees) averaged 1.8 percent of GDP in 2013-2015.
  - Malawi: total telecoms revenues (including license and spectrum fees) averaged 1.3 percent of GDP in 2013-2015.
  - Senegal: mobile phone companies accounted for 20 to 30 percent of corporate income tax between 2005 and 2009.
  - Haiti: over a quarter of the country’s sales tax (excluding customs) came from the telecoms sector in 2014.
- The tax burden on mobile telecommunications increased from 17.4 percent in 2007 to 20.1 percent in 2015 (Deloitte), with the sample size rising from 101 countries in 2007 to 110 countries in 2015.
- In 2016, developed countries had 3.3 times as many cell phone subscriptions as fixed-lined subscriptions, while developing countries had 11.3 times as many.

### Composition of the total tax burden
- The total tax burden on telecoms includes:
  - consumer, corporate, and trade taxes;
  - sector-specific taxes such as CIT surcharges, VAT surcharges, service and handset excises;
  - elevated customs charges on capital equipment;
  - regulatory charges, notably spectrum and/or operator license fees.
- Table of telecommunications-specific tax provisions (sample entries):
  - CIT surcharge: Elevated CIT rate or CIT surcharge on telecoms operators — examples: Cote d'Ivoire, Jamaica, Jordan, Tunisia, Yemen, Zambia.
  - VAT surcharge: Elevated VAT rate on telecom products and/or services — examples: Argentina, Brazil, Jamaica, Mauritania, Sudan.
  - Excises (domestic calls): Specific or ad valorem charge on domestic call minutes — examples: Albania, Bangladesh, Dominican Republic, Ecuador, Greece, Jamaica, Kenya, Malawi, Nepal, Pakistan, Tanzania, Turkey, Uganda, Ukraine, Venezuela, Zambia.
  - Excises (international calls): Specific or ad valorem charge on incoming international calls — examples: Congo-Brazzaville, Gabon, Ghana, Jamaica, Malawi, Senegal.
  - Handsets: Specific or ad valorem charge for domestic and imported handsets — examples: Bangladesh, Botswana, Brazil, Ghana, Nigeria, Lesotho, Syria, Tunisia, Turkey.
  - Connection fees: Specific or ad valorem charge for telephone connection or SIM card — examples: India, Tanzania.
  - Elevated customs duties on operator equipment and handsets — country examples listed.
  - Tax incentives/reduced tax rates — examples: Angola, Congo-Brazzaville, Niger.
- Sources referenced for tax provision inventory: GSMA (2012), Deloitte (2016a, 2016b), dutycalculator.com.

### Sector-specific taxation: rationale and concerns
- Sector-specific taxes in telecoms have become as prevalent as in extractive industries.
- The extractive sector’s higher taxes are justified by economic rents from inelastically supplied, non-renewable resources; the paper seeks to explore whether similar conditions justify higher taxes on telecommunications.
- Concerns:
  - Industry claims of over-taxation.
  - Potential negative growth effects from rising tax burdens.
  - Developing countries often rely on relatively easy-to-administer taxes given lower administrative capacity; this may lead to heavier use of sector-specific taxation compared with developed countries, which can utilize more efficient but harder-to-administer tax bases such as income tax.

### Scope and focus of the paper
- Focuses on taxation of the telecommunications sector in general, with emphasis on cellular telephony because it is the dominant form of communication in the developing world.
- Excludes taxation of media content (e.g., e-banking and e-commerce).
- Examines tax policy and spectrum auctions in detail; does not provide a thorough consideration of non-tax policies such as regulation, state ownership, and competition policy.
- Analytical themes to be addressed include network externalities, growth effects, demand elasticity, industrial organization, and technological and regulatory considerations, and implications for direct and indirect taxes applied to the telecoms sector.

*Source: wp17247 - References (excerpt).*

### conclusion provides some key lessons in designing and implementing a tax system that will help

### wp17247 - conclusion provides some key lessons in designing and implementing a tax system that will help

### II. ECONOMICS OF THE TELECOMMUNICATIONS SECTOR — Network Externalities
- Telecommunications is a network industry in which new users generate positive externalities for existing users by broadening the network.
- Key conceptual points:
  - The value of a communication network lies in its ability to provide access to its members (Economides, 1996). Network expansion provides a free benefit to those already in the network.
  - Positive externalities imply free markets could yield a sub-optimal level of cell phone communications; a subsidy (e.g., for handsets) that lowers the cost of joining a network could be efficient, especially in early network establishment.
  - Common policy responses include cross-subsidizing network expansion to remote areas and “buildout requirements” in license agreements. Universal service funds are sometimes used but are prone to governance problems and under- or mis-utilization.
- Empirical evidence:
  - Doganoglu and Grzybowski (2007): a "700 percent expansion of cell phone communications over 1998-2003" in Germany was largely due to network effects (measured by lagged network size) and less due to a "41 percent price decrease."
  - Grajek (2010) finds similar intra-operator network effects for Poland (1996-2001).
  - Diffusion-model studies (Wu and Chu (2009), Rouvinen (2006)) conclude both network effects (existing penetration) and potential user base significantly impact diffusion.
- Penetration statistics:
  - Ratio of global phone subscriptions (fixed and mobile) to population: "107 percent in developing countries and 165 percent in developed countries."
  - Unique mobile phone subscribers as a percentage of global population: "67 percent worldwide—62 percent in developing countries."
- Policy implication: public policy likely to continue supporting network expansion to underserved areas; care required because universal service funds financed by telecom charges can be inefficient.

### II.B Contribution to Economic Growth
- Mechanisms:
  - Cell phones decrease the cost of acquiring information, reduce transaction costs, create opportunities for additional transactions, and contribute to economic efficiency and growth.
  - Tax measures that decrease or slow down cell phone penetration can yield short-term revenue but harm long-term growth benefits.
- Empirical macro evidence:
  - Waverman, Meschi and Fuss (2005): among low income countries, "10 more mobile phones per 100 people increased per capita GDP growth by 0.6 percentage points—roughly twice as much as in high income countries."
  - Sridhar and Sridhar (2007): "an increase of 1 percent in mobile penetration increased GDP by 0.01 percent"; an average cumulative growth in cell phone use of "248 percent increased GDP by 2.48 percent" over the study period.
  - Lee, Levendis and Gutierrez (2012): cell phone penetration had significant impact on growth recently (2000-2006) in 44 sub-Saharan countries; impact stronger where landline penetration is lowest.
  - Andrianaivo and Kpodar (2012): "a 10-percentage point increase in the mobile penetration rate increases the growth rate by 0.6 percentage points." They also find the price of a "3-minute conversation is negatively related to growth."
- Microeconomic evidence (case studies and reviews):
  - Documented welfare sources: (1) access to and use of information; (2) productive efficiency; (3) new jobs for mobile-related services; (4) reduced exposure to shocks and risk; (5) delivery platform for key development services.
  - Examples:
    - Jensen (2007): reduced price dispersion, lower prices, higher profits, elimination of waste after cell phone introduction.
    - Muto and Yamano (2009): mobile expansion associated with greater market participation of remote producers of perishable goods.
    - Acker (2010): introduction of cell phones in Niger between 2001 and 2006 lowered price dispersion by "10 to 16 percent" on grain markets.
    - Acker and Fafchamps (2015): "6 percent" reduction in producer price dispersion in Niger following cell phone introduction.
  - Strong link between cell phone use and financial inclusion; expansion of e-banking also depends on regulation, inter-operability, retail agents, and institutions.
- Policy nuance:
  - Positive macro and micro evidence generally suggests a positive causal relationship between cell phone use and growth.
  - This does not necessarily imply externalities requiring government intervention; transactions may be properly priced.
  - Evidence cautions against taxes that deter business and telecommunications access—see discussion on excise vs. VAT surcharge.

### II.C Demand Elasticity
- Rationale analyzed: telecommunication excises are sometimes justified by inelastic demand offering a good "tax handle."
- Empirical elasticity evidence (diverse and time-varying):
  - Commerce Commission (2003) review of 13 papers:
    - Network access: "-0.06 to -0.54"
    - Mobile originated calls: "-0.09 to -0.8"
  - Waverman, Meschi and Fuss (2005): "-1.5" (aggregate production function for 38 low- and lower middle income-countries)
  - Garbacz and Thompson (2007): connection charges: "-0.37 to -0.029"; monthly price: "-1.268 to -0.195"
  - Dewenter and Haucap (2008) Austria traffic:
    - Short-term, businesses: "-0.33"
    - Short-term, private consumers: "-0.14"
    - Long term, businesses: "-0.74"
    - Long term, private consumers: "-0.37"
  - Gasmi et al. (2009) South Africa (sample of 6936 individuals in 2005): "Between -3.8 and -1.3"
  - Caves (2011) USA: "-2.1"
  - Koutroumpis et al. (2011) Greece: "-1.6"
  - Hakim and Neaime (2014) MENA: "Between –1.2 and -1.0"
  - Karacuka et al. (2011) Turkey:
    - Short term, pre-paid: "-0.20"
    - Short term, post-paid: "-0.36"
    - Long term, pre-paid: "-0.33"
    - Long term, post-paid: "-0.72"
- Conclusion: diversity of elasticity estimates is striking; network access elasticity tends to be generally lower than usage elasticities. Whether inelasticity justifies excise taxation is discussed later in tax policy sections.

### II.D Technology, Industry Structure, and Market Power
- Historical shift:
  - Fixed-wire telephony was treated as a "natural monopoly" due to high fixed costs and economies of scale; prices aligned with average cost plus normal return (Averch and Johnson, 1962).
  - From the 1980s onward, liberalization, privatization, unbundling, and the spread of cellular networks and VoIP transformed the sector into a more contested market.
- Constraints on effective competition:
  - Entry costs remain substantial despite lower fixed costs for cell networks than fixed-line networks (GSMA, 2008).
  - Limited electromagnetic spectrum constrains the number of potential service providers. Frequencies in the "800-900 megahertz (MHz) range" are best suited for cell communications; frequencies up to "2+ gigahertz (GHz)" are sometimes used.
  - Typical country market structure: "Most countries have 2-4 providers, while some retain a monopoly structure"; "Less than one fifth of countries have at least 5 providers."
  - Li and Lyons (2012) identify "5 providers and more" as the threshold for full effective competition.
- Evidence of rents and profitability:
  - Limited competition facilitates implicit or explicit collusion and higher service prices.
  - US telecom sector profitability since 2009: average return on assets of large US telecom service companies exceeded that of large-capitalization companies in general by "70 percent"; return on investment and return on equity were "16 percent and 15 percent higher than the market, respectively."
  - Boniecki and others (2016) for Middle East and Africa: dominant operators' EBITDA-to-revenue ratios "41-47 percent", second-largest "30-37 percent", third and fourth "15-25 percent".
- Technological and organizational changes (VoIP, data-based communications, virtual mobile network operators) can shift rents and competition—outcomes depend on regulatory and licensing policies.

### II.E Regulation
- Regulatory policy affects magnitude of rents via spectrum licensing, license fees, and price regulation.
- Spectrum management and market depth:
  - Spectrum policy determines the number of mobile telecom licenses; country size positively influences the number of operators and efficiency (Symeou, 2011).
  - Market concentration data: average market share of second providers is less than "one third"; third and fourth providers less than "20 percent" and "10 percent", respectively.
- Regulatory interventions:
  - Price regulation may still be necessary where incumbents control key infrastructure or have monopoly over connections to their network.
  - Buildout requirements and universal service funds can extend coverage but reduce operator profitability; universal service funds may be inefficiently deployed.
  - Regulatory charges include spectrum and license fees (often substantial) and smaller fees for monitoring; spectrum and license charges have significant impact on profitability.
- Governance considerations:
  - Regulatory charges should finance regulatory costs; charges above cost recovery are taxes (or subsidies if below cost).
  - Fiscal authority should be concentrated in the treasury; residual revenue from spectrum licenses or other fees raised by regulators should flow to the treasury to avoid divergence of interest and "overgrazing the commons."

### III. TAX POLICY — General Considerations
- Core tax system objectives: finance public goods, ensure efficiency, equity, and simplicity — few taxes with low rates on wide, uniform bases.
- Exceptions that may justify sectoral differential treatment:
  - Industries generating economic rents (e.g., extractive industries) can be efficiently taxed without distorting investment incentives.
  - Presence of externalities: negative externalities taxed, positive externalities subsidized.
  - Administrative infeasibility of standard taxes for particular activities may justify second-best arrangements.
- Telecom-sector specifics:
  - Technological and regulatory barriers to entry, limited spectrum, and small market sizes can generate rents—suggesting potential suitability for some form of rent taxation.
  - Collection of profit-based taxes (CIT, rent taxes) may be difficult due to profit-shifting by MNEs; in low administrative capacity contexts, simpler instruments like excises may be used despite greater distortion.
  - Positive externalities suggest cross-subsidies for network extension and avoiding taxes that specifically limit access.
- Public finance principle: taxes on productive inputs should be avoided; customs duties on capital equipment should ideally be zero.

### III.B Spectrum License Fees
- Form and duration:
  - Spectrum license fees are usually fixed payments levied up-front and/or periodically; typical license duration "about 15 years."
- Allocation methods and effects:
  - Competitive auction preferred since 1990s; auctions allocate licenses to efficient providers and procure revenue.
  - Theoretical expectation: bidders pay up to the rents from holding the license.
- Empirical and practical findings:
  - Some studies (Morris (2005), Kwerel (2000), Bauer (2003)) find no evidence license fees impact service pricing—supporting the view that auctioned license fees can function like a rent tax if not insolvency-inducing.
  - More recent work indicates auctions may affect pricing and penetration: Janssen and Karamychev (2009) show less risk-averse bidders may price higher post-auction; Haan and Toolsema (2011) show debt-financed bids affect post-auction pricing; Kuroda and del Pilar Baquero Forero (2017) find penetration rates are lower in countries that conduct auctions.
- Risks and design issues:
  - Maximizing auction revenue can create perverse incentives: limiting size/number of licenses can reduce competition and raise consumer prices.
  - Collusion risk among a small number of large multinational bidders; faulty auction design and excessive reserve values can cause auction failure.
  - Historical undervaluation of licenses in some markets led to ex-post aggressive taxation when governments later sought to recoup concession value.
- Valuation complexity:
  - License valuation depends on spectrum amount, market size and depth, license terms (service types, technological requirements, price caps, build-out, government equity stakes, transfer rights).
  - Ex-ante fixed fees put all financial risk on providers and may lead to shaded bids.
  - Alternative fee structures:
    - Ex-post profit- or gross-revenue sharing (royalties) distribute risk more evenly and can create incentives for regulators and operators to expand spectrum allocation and investment.
    - Kwon, Lee and Oh (2010) favor royalties over ex-ante fixed fees.
- Policy recommendation: improve government capacity for accurate spectrum valuation; consider auction design, reserve values, and risk-sharing fee structures.

### III.C Rent Taxes
- Concept:
  - Resource rent taxes (RRTs) tax profits above the “normal rate of return” for an industry.
  - Example: if normal rate of return is "15 percent", an equity-financed investment of US$100 yielding profit US$25 generates a rent of US$10; CIT base is US$25, RRT base is US$10.
- Application to telecoms:
  - Authors note RRTs have not yet been applied to the telecom industry (to their knowledge).
- Design features of RRTs:
  - Interest expense typically disallowed; investment fully expensed rather than depreciated.
  - Losses carried forward with interest to preserve real value of expensing.
  - Revenue profile of RRT is more back-loaded than CIT, entailing higher public revenue risks.
  - RRTs in extractive industries typically have statutory rates of "15-30 percent" and are applied in addition to—not instead of—the standard CIT.  

*Source: wp17247 - conclusion provides some key lessons in designing and implementing a tax system that will help (PDF).*

### Box 1. Rent Taxation Options

### Box 1. Rent Taxation Options

### Rent-based tax designs
- Brown Tax or R-based cash flow tax
  - State acts as a passive investor, meeting its share of all net negative cash flows by direct cash payment at the same percentage as the tax rate, and taxing all net positive cash flows at the same rate.
  - Narrows distribution of possible outcomes but does not change the mean expected return; any tax paid is thus a tax on rent.
  - Accounting and tax depreciation do not feature: all capital is immediately expensed, calculations are on cash flows.
  - The Australian RSPT proposal of 2010 approximated this tax (Henry, and others 2010).
- Resource rent tax
  - Replicates many features of the Brown tax, but the investor receives an annual uplift on accumulated losses until these are recovered rather than direct cash payments by the state.
  - The uplift rate should be set at the minimum required rate of return for the investor; if losses can be completely offset against profits before project end, the tax is neutral and the uplift factor should come close to a risk-free interest rate.
  - Calculations use cash flows, not book or tax depreciation.
- Variable Income Tax (VIT)
  - Uses the CIT base but varies the rate according to the ratio of profits to gross revenues.
  - Developed in the gold mining industry of South Africa; effective tax rate may be lower or higher than standard CIT. Variant modeled here permits only a higher rate.
  - Relatively simple but may introduce distortions if high accounting profit occurs early in project life before required return earned.
- Tax surcharge on cash flow
  - Adjust accounting profit base by adding back depreciation and interest and deducting any capital expenditure in full to yield net cash flow in the year; this can form the base for a surcharge.
  - Instead of annual uplift for losses, the rate could be set low to imply compensation, or a simple uplift (investment allowance) could be added at the start.
  - Example: U.K. North Sea petroleum projects — CIT surcharge rate from 2011/12 is 32 percent, in addition to normal CIT.
  - Could distort by taxing early revenue, but effect likely small.
- Allowance for Corporate Capital (ACC) surcharge scheme
  - Permits an annual uplift on the balance of undepreciated capital assets on the books rather than converting to cash flow.
  - Actual interest paid is not deductible; ACC creates neutrality between debt and equity financing and should make investor indifferent to rate of tax depreciation.
  - Norway uses this scheme to tax rents from hydropower projects, with a rate of 30 percent added to the CIT rate (24 percent in 2017 and then 23 percent in 2018).

### Efficiency, neutrality, and administrative considerations
- Rent-raising features and neutrality
  - RRTs tax only rent and not the normal return to investment, making them less distortive than the CIT.
  - Well-structured rent taxes should have no impact on marginal investment, production, or pricing; they should not impede network access.
  - If rent taxes increase average tax rate on investment (ratio of total taxes paid to total pre-tax profits), they may affect entry decisions and reduce valuation of spectrum licenses.
- Equity and timing
  - Rent taxes are arguably more equitable than spectrum auctions because they are based on actual rents earned rather than expected future rents at bidding.
  - If an operator earns no rents, liability under a rent tax would be zero.
  - Rent taxes are less vulnerable to collusion than weakly structured auctions.
  - Rent taxes are most easily introduced at the outset of an investment project and should ideally be timed with licensing and investment cycles.
- Vulnerabilities and administrative costs
  - Like the CIT, RRTs are vulnerable to accounting manipulation and aggressive transfer pricing to shift profits to lower-tax jurisdictions.
  - Designing and implementing a new tax entails significant administrative costs that must be weighed against potential revenue and efficiency gains.
  - Mintz (2017) notes that where both an RRT and the standard CIT are applied, their interaction can interfere with the neutrality of the RRT with respect to investment decisions.

### Corporate Income Tax (CIT) implications
- Use of higher CIT rates
  - Several countries (including Jamaica, Côte-d’Ivoire, Jordan and Yemen) subject telecom companies to higher CIT rates to capture perceived above-normal profits.
  - In a short-run, static context, a proportional profit tax is unlikely to directly affect consumer pricing if operators maximize profits.
- Distortionary effects
  - Unlike rent taxes, the CIT taxes the normal return to capital, raising the cost of capital and depressing investment; output is therefore likely to be lower and prices higher under an elevated CIT rate.
  - Davidson and Martin (1985) demonstrate higher CIT rates may augment oligopolistic collusion by making more restrictive equilibria possible.
- Profit shifting and mobility
  - Imposing a CIT surcharge increases incentives for MNEs to shift profits across borders through aggressive tax planning.
  - Heckemeyer and Overesch (2013) find an average semi-elasticity of multinational corporate profits to the CIT rate of -0.8 (i.e., a 1 percentage-point higher CIT rate will reduce reported profits of an affiliate of a foreign company by 0.8 percent).
  - Transfer of telecom licenses indirectly through offshore mergers and acquisitions can be used to avoid capital gains taxes (example cited: Vodafone/Hutchison case avoided $2.6 billion in capital gains tax).
- Comparative assessment
  - Spectrum auctions, rent taxes, and CIT surcharges can all tax telecom rents; each has strengths and weaknesses:
    - Spectrum auctions reflect expected rather than actual rents and are vulnerable to collusion.
    - Rent taxes are likely the most efficient and equitable but defer revenues.
    - CIT surcharges are more distortive but administratively less costly if CIT systems already exist.
  - Choice depends on administrative capacity and the stage of the licensing and investment cycle.

### Telecom excises: rationale, effects, and evidence
- Rationale for excises
  - Profit-based taxes on capital-intensive industries yield revenues with a lag; many countries introduce telecom excises to ensure significant up-front revenue.
  - Excises are relatively easy to administer and important where administrative capacity is low.
  - Excises function similarly to royalties in extractive industries: levied as percentage of output value or fixed charge per unit, they raise revenue from start of production and are less vulnerable to tax evasion than CIT or rent taxes.
- Economic distortions and externalities
  - Excises raise price of a good relative to others and suppress consumption; effect is smaller the less own-price elasticity of demand.
  - Telecom services typically have positive externalities and are not standard excisable goods (unlike alcohol, tobacco, petroleum).
  - Because excises tax gross revenues, they risk rendering marginal projects unprofitable and reducing investment and output.
- Excises on domestic calls and country evidence
  - Telephony excises may be specific (fixed charge per minute) or ad valorem (percentage of service charges); both are proportional to output and gross revenues and thus roughly proportional to profits for a given price level.
  - Malawi: domestic excises account for almost half of total fiscal and regulatory revenues generated by the telecoms sector, which averaged 1.3 percent of GDP in 2013-2015.
  - Jamaica: telephone excises plus the VAT surcharge account for more than one third of total telecom revenues, which averaged almost 1.8 percent of GDP in 2013-2015.
  - Jamaican VAT on telecom service is 25 percent, vs. the standard VAT rate of 16.5 percent; the Jamaican VAT surcharge represents 34 percent of VAT revenue, or 0.22 percent of GDP.
- Price incidence and market structure effects
  - An excise generally increases telecom prices for consumers and businesses; incidence depends on relative price elasticities of supply and demand.
  - Over the medium term, disinvestment due to reduced profitability can reduce capacity and drive up consumer prices, shifting burden onto consumers.
  - Excises can accelerate substitution to VoIP; many countries have extended excise base to include internet data to forestall base erosion (e.g., Turkey, Guinea).
  - Where competition is limited, service excises may increase market power of dominant firms by driving out higher-cost operators; excises can facilitate collusion and lead to “overshifting” of the tax to consumers.
- Guinea case study: impact of a new excise
  - Tax introduced: 1 GNF per second (so 60 GNF per minute, or nearly 15 percent of the average price per minute in Q2-2015).
  - Traffic decreased by 16.1 percent in Q3-2015.
  - Quarterly revenue from the tax is roughly 12 M USD (exchange rate of 10,000 GNF/USD).
  - SMS component of the same tax had an even more dramatic impact.
- Trade-offs for policy design
  - Moderate royalties/excises can play an important role as part of a balanced mix of tax instruments to share fiscal and economic risks between government and investors.
  - Excises are less effective than rent taxes or CIT at capturing rents, and they distort consumption and business inputs, potentially reducing productivity and welfare.
  - Policy choice among auctions, rent taxes, CIT surcharges, and excises should reflect administrative capacity, timing in licensing/investment cycle, and the need for up-front revenue versus efficiency.

*Source: wp17247 - Box 1. Rent Taxation Options (IMF).*

### Introduction of the

### Introduction of the SMS component of the TCT

### Specific versus ad valorem excises: welfare and price effects
- Literature generally concludes that ad valorem excises dominate specific excises in terms of welfare effects in markets with limited competition.
- Under oligopoly, equilibrium prices will be lower under an ad valorem excise than under an equivalent specific excise.
- Mechanism:
  - An ad valorem excise flattens the effective demand curve that operators face, while a specific excise shifts the curve downward.
  - Facing a more elastic effective demand curve nudges operators’ behavior toward what it would be in a perfectly competitive market.
  - With an ad valorem excise, part of the cost of a price cut is borne by the government in terms of lower tax revenues, so producers have more incentive to reduce prices.
- Practical advantage: Ad valorem excises are more straightforward to extend to internet data, since internet service has different consumption units than telephony (megabytes vs. calling minutes).

### Excises on international calls: prevalence, impacts, and distortions
- Recent trend: In the last 15 years, some governments have taxed international incoming calls at rates that far exceed domestic call excises, motivated by the view that the tax falls on foreigners and by short-term revenue needs.
- Examples and figures:
  - Burundi raised the rate of its termination fee to 247 percent of the underlying charge (GSMA, 2014).
  - In Jamaica, the excise on domestic calls is about US$0.003 per minute, while the rate on international calls is US$0.075 per minute.
- Empirical evidence on traffic elasticity:
  - When Ghana and Gabon imposed taxes that raised incoming call prices by 58 and 82 percent, call volume dropped by 27 and 57 percent, respectively (GSMA, 2014; OECD, 2014).
- Technological factors increasing tax base elasticity:
  - Shift toward internet calling using VoIP.
  - Use of illegal “SIM boxes” which use local SIM cards to transform international internet-carried calls into local phone calls.
  - GSMA (2014) estimate: in Ghana, 10 percent of international calls were re-routed through SIM boxes.
- Political and economic costs:
  - Incoming call excises often fall heavily on a country’s own diaspora and business partners, generating vocal opposition.
  - Economic costs include higher communication costs for regional calls and reduced CIT revenues and remittances.
  - Regional policy response example: May 2014 agreement among Kenya, Rwanda, Burundi, Uganda, and South Sudan not to tax each other’s outgoing calls.
- Policy recommendation:
  - International call excises should be limited to the same rate as domestic telephone excises to avoid distortions and eliminate the incentive to use SIM boxes.

### International call termination process (summary of Box 2)
- Call routing steps (example: France to Democratic Republic of the Congo):
  - Caller’s handset → local tower (cell) → switching point → submarine fiber optic cable (e.g., Africa Coast to Europe, “ACE” cable) → arrival in destination country → transfer back to cell network/land line → end user.
  - The ACE cable is owned by a consortium led and managed by the French operator Orange.
- Tax implementation:
  - Taxes on international incoming calls are generally imposed through a minimum price to be charged for connecting (“terminating”) the call to the end user’s operator in the destination country; the tax is remitted to the destination country government as a share of that minimum price.
- Interaction with internet data and avoidance:
  - Internet data can travel through the ACE cable without being subject to the telecom excise if not taxed; a VoIP call may reach a recipient without being subjected to the incoming call tax.
  - Illegal SIM boxes can convert VoIP/internet traffic into local calls, subjecting them only to domestic call excises (not higher international excises).

### Excises on network access (handsets and connection charges)
- Some countries impose excises on up-front costs of establishing phone service, including handsets and initial connection charges (e.g., SIM cards).
- Harms of such excises:
  - Compared to a service excise that raises the same revenue over time, excises on phone acquisition and initial connection are particularly harmful because they may altogether prevent access—especially in low-income countries—reducing the size and value of the telecom network for all users.
  - Presence of positive network externalities may indicate initial access should be subsidized rather than taxed.
  - High excises or customs duty on small, portable items like cell phones can promote smuggling.
  - Taxes on handsets and initial connection seldom raise significant revenues and should be eliminated.

### Value-added Tax (VAT) on telecoms: design issues and practical adaptations
- Most countries apply a VAT to telephone services at the standard rate.
- Differences from excises:
  - Due to input VAT crediting, the VAT is not a cost for registered VAT payers, so production distortion created by telecom excises is alleviated for formal-sector companies above the VAT threshold.
  - Informal companies still experience the higher VAT rate as a cost.
  - To raise the same revenue from a VAT surcharge as from an excise, the surcharge would have to be higher to offset input crediting, increasing price distortion for other users.
- Risks of VAT surcharges and dual rates:
  - A VAT surcharge could complicate tax policy and administration.
  - Dual VAT rates are difficult to contain to a single industry and reduced VAT rates tend to “creep” to other goods and services.
  - The IMF has consistently supported a single-rate VAT.
  - If telecom expenses generate excess VAT credits, refunds would be required, exposing weak points in VAT administration in many low- and middle-income countries.
- Place-of-consumption and international call VAT complexities:
  - Properly and consistently implementing VAT on international calls can be complex; cross-country confusion on the place of consumption may create departures from basic VAT principles, double taxation, or no taxation.
  - Some countries have applied VAT to net international revenue (access fees charged to, minus access fees paid to foreign carriers), effectively applying VAT to net exports; removing such practices has met with fierce resistance in low-income countries given revenue implications.
- Practical adaptation example: VAT pre-payment and calling cards (Box 3; Senegal example)
  - Example figures preserved exactly as in source:
    - Final consumer price: 1000 CFA Francs (FCFA) with an 18 percent VAT.
    - Pre-VAT price: 1000/1.18 = 847.46 FCFA.
    - VAT amount: 152.54 FCFA.
    - Operator might sell the card for pre-VAT price of 750 FCFA to the bulk distributor, to which the VAT on the final retail price is added, for a total of 902.54 FCFA.
    - Bulk distributor might then resell the card VAT-free to the street vendor for 960 FCFA, who then resells it to the final consumer for 1000 FCFA.
    - Effective VAT rate on the first transaction: 152.54/750, i.e., 20.3 percent.
  - Administrative rationale: collecting VAT at the operator level on the nominal amount of the recharge avoids daunting administrative difficulties along the distribution chain and addresses the need to sell calling cards at a discount to allow profits for downstream distributors.

*Source: IMF working paper excerpt (wp17247).*

### Box 4. Prepaid cards, and international calls: who pays the VAT?

### Box 4. Prepaid cards, and international calls: who pays the VAT?

### Principles and practical challenges
- VAT on telephone calls should apply at the point of consumption (destination principle – hence exports are zero-rated).
- All VAT paid on inputs should be offset against VAT collected on sales, as per the credit-invoice method (except for non-registered entities – i.e., the final consumers).
- In practice, a conversation involves two individuals and potentially many operators; payment systems may differ (e.g., caller pays vs. shared payment), making determination of the point of consumption and VAT liability problematic.
- A common agreement on the application of VAT for cellular communications reduces administrative difficulty but has significant administrative implications.
- Recent revisions of the EU Directive on the Value-added Tax recommend that the country of origin of the calling SIM card be considered the place of consumption. (See: Council Implementing Regulation (EU) No 1042/2013 of 7 October 2013 amending Implementing Regulation (EU) No 282/2011 as regards the place of supply of services. See also: OECD (2017), International VAT/GST Guidelines, OECD Publishing, Paris.)

### Implications of treating the SIM card country of origin as place of consumption
- The origin-based rule simplifies VAT treatment by deeming calls consumed in the country of the calling SIM card, even if the call physically takes place elsewhere.
- This norm has numerous implications in many low- and middle-income countries where VAT-able prepaid card / “caller pays” systems are in place and where VAT treatment between operators is often disputed, especially for international calls.
- Administrative difficulties under the destination principle include the need for the caller to register and pay VAT where the call is consumed or for the home operator to transfer VAT to the foreign operator at the foreign rate.
- The origin-based norm notably implies that VAT cannot be applied to the net income from transactions between Home and Abroad operators, as current practice in some countries.

### VAT treatment by scenario (country of originating SIM card / locations / call type / VAT treatment)
- Home / Home / Home / Base case
  - Originating Home SIM pays VAT
  - The caller’s operator pays a VAT to the receiver’s operator and credits it as input VAT
- Home / Home / Abroad / Outgoing
  - Originating Home SIM pays VAT
  - Home operator pays access fee to foreign operator (import of service) and reverse charges the VAT (charges the VAT to itself and remits it to Home tax authorities) at Home rate. This input VAT can be credited
- Home / Abroad / Home or Abroad / Outgoing roaming
  - Although the service is consumed Abroad, it is deemed consumed at Home, because the call is from a Home SIM card. Home VAT therefore applies
  - Home operator pays access fee to foreign operator (import of service) and reverse charges the VAT (charges the VAT to itself and remits it to Home tax authorities) at Home rate. This input VAT can be credited
- Abroad / Home / Home or Abroad / Incoming roaming
  - Although the service is consumed at Home, it is deemed consumed Abroad, because the call is from an Abroad SIM card.
  - The foreign operator pays an access fee to the Home operator, who exports a service, and the transaction is zero-rated
- Abroad / Abroad / Home / Incoming
  - The Home receiver does not pay the VAT
- Abroad / Abroad / Abroad / NA
  - NA

### Broader fiscal and regulatory implications (input taxes, user fees, and tax incentives)
- Input taxes and user fees
  - Countries sometimes levy charges on telecom inputs to capture a share of sectoral profits; these take the form of elevated customs duties on telecoms capital equipment and/or user fees on public inputs (e.g., charges for use of public infrastructure, land for antennas, issuance and management of phone numbers, industry monitoring (minute counts)).
  - It is economically efficient to charge for public services when cost can be calculated reasonably and an exclusive user identified; regulatory services often meet this criterion.
  - To the extent that fees exceed the cost of providing a service, they comprise an input tax.
  - In many telecom cases the public service is ill-defined or non-existent, or its cost is overstated or reflects inefficient public service provision.
  - Input taxes can worsen limited competition: taxing inputs in an oligopolistic industry alters equilibrium prices, output and profits (Katz and Rosen (1985)). If marginal costs are nondecreasing in the input tax, it will increase market prices and reduce output, and the input tax can be overshifted to the consumer.
  - Since input taxes distort production and increase prices, they should be avoided in favor of more efficient types of taxes. Regulatory fees should be restricted to covering the cost of service provision.
- Tax incentives
  - Governments often offer telecom operators tax incentives available to other sectors (e.g., reduced taxes on profits, investment incentives, customs duty waivers). Examples noted: Niger offered an initial 8-year holiday for CIT and some other taxes to four operators; the Central African Republic has similar concessions.
  - Many developing-country investment-code incentives are ill-advised, sacrificing scarce revenues on investment that would have taken place without a special incentive.
  - Customs waivers on capital goods can be justified on productive efficiency grounds, but a preferable policy is to specify a zero tariff on capital goods in the customs code to minimize discretion.
  - Telecom operator licenses confer exclusive rights to exploit “locational rents” from selling services into the domestic market; extending investment incentives to telecoms is difficult to justify and likely to cause revenue losses with minimal impact on investment.
  - Network externalities may justify lenient terms for establishing networks in new areas; initial cellular licenses were often granted at no charge.
  - Telecommunications is not labor-intensive and is arguably not an appropriate candidate for incentives aimed at boosting regional employment; ensuring coverage in remote regions is better achieved via buildout requirements and universal service funds rather than tax breaks.

### Conclusions and policy guidance (pecking-order approach)
- Taxes on the telecoms sector have proliferated, increasing uncertainty and compliance costs (elevated customs duties, diverse regulatory fees, elevated CIT and VAT rates, telephone call excises).
- Where rents from limited competition cannot be addressed through regulatory measures, they may justify a higher level of taxation. However, many fiscal instruments applied to telecoms do not target rents well and distort markets.
- A “pecking order” of tax instruments: rely on taxes that best capture rent and resort to consumer-affecting measures only as a fallback.
  - Auctions: capture rents without distorting investment and prices; favored for countries with regulatory capacity. Risks: auctions are based on expected rents (may leave windfalls or strain operators), collusion/governance issues, impossibility where long-term licenses already awarded, and risk of encouraging monopolistic behavior if revenue-maximization is over-emphasized.
  - Rent tax or CIT surcharge: capture a portion of telecom rents with minimal impact on prices; more equitable and flexible than auctions because based on actual profits. Weaknesses: earnings manipulation and transfer pricing; CIT raises cost of capital (reduces investment), rent taxes lead to revenue deferral and higher administrative cost.
  - Taxes on consumption (service excises): imperfect for targeting rents and raise prices/restrict access; simple to design and administer and hard for firms to avoid. If used, ad valorem excises are preferable to specific excises because they are less likely to be over-shifted to consumers. A VAT surcharge may be preferable to an ad valorem excise since it does not burden formal businesses but may necessitate a higher rate on consumers and complicate administration. Excises on handsets and initial connection should be avoided to protect network expansion externalities.
  - Avoid taxes on producer inputs (customs duties, excessive public user fees) as they do not target rents effectively and distort markets.
  - Avoid tax incentives for telecoms as unnecessary to attract investment to exploit the local consumer market; treat telecom taxes the same as for other sectors except where clear, limited justifications exist (e.g., buildout requirements, universal service funds).
- A balanced tax regime sharing fiscal and economic risks between investors and government is advisable. Rapid technological change requires caution and adaptability in tax system design; persistence and allocation of telecom rents will depend on regulatory regimes and market dynamics that encourage effective competition.

*Source: Box 4. Prepaid cards, and international calls: who pays the VAT? (wp17247).*

### REFERENCES

### REFERENCES

### Taxation and Fiscal Regimes
- Daniel, P., P. Harris, O. Luca, and C. Nakhle. 2011. Sierra Leone: Fiscal regimes for extractive industries – A preliminary review. Fiscal; Affairs Department, International Monetary Fund.
- Deloitte. 2007. “Global Mobile Tax Review, 2006-2007.”
- Deloitte. 2012. “Mobile telephony and taxation in Turkey.”
- Deloitte. 2015. “Digital Inclusion and Mobile Sector Taxation.”
- IMF, OECD, UN, and World Bank Group. 2017. “The Taxation of Offshore Indirect Transfers—a Toolkit,”
- IMF, OECD, UN and World Bank. 2015. Options for Low Income Countries’ Effective and Efficient Use of Tax Incentives for Investment: A report to the G-20 Development Working Group. October 2015.
- Khan, T. 2007. Excise Taxation of Telecom Goods and Services in Developing Countries. International Monetary Fund. Unpublished manuscript.
- Mintz, J. 2017. “Taxes, Royalties and Cross-border Resource Investments.” In International Taxation and the Extractive Industries, edited by P. Daniel, M. Keen, A. Swistak and V. Thuronyi, Routledge, London, 306-331.
- Schatan, R. 2012. “Tax Minimizing Strategies and the Arm’s-Length Principle.” Tax Notes International, January 9: 121–26.
- Liu, L., and R. Altshuler. 2013. “Measuring the Burden of the Corporate Income Tax under Imperfect Competition” National Tax Journal 66(1): 215-238.
- Heckemeyer, J. H.; and M. Overesch. 2013. “Multinationals' profit response to tax differentials: Effect size and shifting channels, ZEW Discussion Papers, No. 13-045.
- Boadway, R., and M. Keen. 2012. “Principles of Resource Taxation for Low-income Countries,” In The Taxation of Petroleum and Minerals: Principles, Problems and Practices, edited by P. Daniel, M. Keen, and C. McPherson, 13-74. London and New York: Routledge.
- Land, B. C. 2010. “Resource Rent Taxes: A Re-appraisal.” In The Taxation of Petroleum and Minerals: Principles, Problems and Practices, edited by P. Daniel, M. Keen, and C. McPherson, 241-262. London and New York: Routledge.
- Morris, A. C. 2005. “Spectrum Auctions: Distortionary Input Tax or Efficient Revenue Instrument?” Telecommunications Policy, 29(9-10): 687-709.

### Spectrum Allocation, License Fees, and Auctions
- Cramton, P. 2002. “Spectrum Auctions.” In Handbook of Telecommunications Economics, Edited by M. Cave et al., Amsterdam, Elsevier Science, 605-639.
- Hazlett, T. 2008. “Property Rights and Wireless License Values.” Journal of Law and Economics. 51(3): 563-598.
- Hazlett, T., R. Munoz, and D. Avanzini. 2011. “What Really Matters in Spectrum Allocation Design.” George Mason University Law and Economics Research Paper Series 11-48.
- Haan, M. A., & Toolsema, L. A. 2011. License auctions when winning bids are financed through debt. Journal of Industrial Economics, 59(2), 254–281.
- Janssen, M. C. W., & Karamychev, V. A. 2009. Auctions, aftermarket competition, and risk attitudes. International Journal of Industrial Organization, 27(2), 274–285.
- Kwerel, E. 2000. Spectrum Auctions Do Not Raise the Price of Wireless Services: Theory and Evidence. Federal Communications Commission, Office of Plans and Policy, Washington DC.
- McMillan, J. 1995. “Why Auction the Spectrum?” Telecommunications Policy 19(3):191-199.
- Morris, A. C. 2005. “Spectrum Auctions: Distortionary Input Tax or Efficient Revenue Instrument?” Telecommunications Policy, 29(9-10): 687-709.
- Prasad, R. 2015. “The Production Function Methodology for Estimating the Value of Spectrum.” Telecommunications Policy 39(1):77-88.
- Kuroda, T., and M. del Pilar Baquero Forero. (2017). “The effects of spectrum allocation mechanisms on market outcomes: Auctions vs. Beauty contests.” Telecommunications Policy, forthcoming.
- Bazelon, C., and G. McHenry. 2013. “Spectrum Value.” Telecommunications Policy 32:737-747.
- Bauer, J. 2003. “Impact of License Fees on the Prices of Mobile Voice Service.” Telecommunications Policy, 27: 417-434.
- Kwon, Y., and B. Kim. 2012. “Royalties vs. Upfront Lump-Sum Fees in Data Communication Environments.” Telecommunications Policy, 36(2): 127-139.
- Kwon, Y., J. Lee, and Y. Oh. 2010. “Economic and Policy Implications of Spectrum License Fee Payment Methods.” Telecommunications Policy, 34(3): 175-184.
- Hazlett, T., R. Munoz, and D. Avanzini. 2011. “What Really Matters in Spectrum Allocation Design.” George Mason University Law and Economics Research Paper Series 11-48.

### Mobile Phones, Financial Inclusion, and Economic Development
- ADB. 2013. “Financial Inclusion and Integration through Mobile Payments and Transfer.” African Development Bank.
- Andrianaivo, M., and K. Kpodar. 2012. “Mobile Phones, Financial Inclusion and Growth.” Review of Economics and Institutions 3(2):1-30.
- Aker, J. 2010. “Information from Markets Near and Far: Mobile Phones and Agricultural Markets in Niger.” American Economic Journal: Applied Economics 2:46-59.
- Aker J. C., and M. Fafchamps. 2015. “Mobile Phone Coverage and Producer Markets: Evidence from West Africa.” World Bank Economic Review 29(2)262-292.
- Aker, J. C., and I. Mbiti. 2010. “Mobile Phones and Economic Development in Africa.” Journal of Economic Perspective 24(3):207-232.
- Asongu, S. A. 2015. “The impact of mobile phone penetration on African inequality.” International Journal of Social Economics 42(8):706-716.
- Donner, J. 2008. “Research Approaches to Mobile Use in the Developing World: A Review of the Literature.” The Information Society: An International Journal 24(3):140-159.
- Duncombe, R. 2016. “Mobile Phones for Agricultural and Rural Development: A Literature Review and Suggestions for Future Research.” European Journal of Development Research 28(2):213-235.
- Goyal, A. 2010. “Information, Direct Access to Farmers and Rural Market Performance in Central India.” American Economic Journal: Applied Economics 2:22-45.
- Jensen, R. T. 2007. “The Digital Provide: Information, Technology, Market Performance, and Welfare in the South Indian Fisheries Sector.” Quarterly Journal of Economics 122(3):879-924.
- Muto, M. T. Yamano. 2009. “The Impact of Mobile Phone Coverage Expansion on Market Participation: Panel Data Evidence from Uganda.” World Development 37(12):1887-1896.
- Waverman, L., M. Meschi, and M. Fuss. 2005. The impact of telecoms on economic growth in developing countries. Moving the debate forward. The Vodafone Policy Paper Series (2): 10-23.
- Wu, F.-S., and W.-L. Chu. 2009. “Diffusion Models of Mobile Telephony.” Information Economics and Policy 22:130–143.
- Rouvinen, P. 2006. “Diffusion of Digital Mobile Telephone: Are Developing Countries Different?” Telecommunications Policy 30(1):46-63.

### Telecommunications Infrastructure, Productivity, and Economic Growth
- Cronin, F. J., E. B. Parker, E. K Colleran, and M. A. Gold. 1991. “Telecommunications Infrastructure and Economic Growth: An Analysis of Causality.” Telecommunications Policy 15(6):529-535.
- Datta, A., and S. Agarwal. 2004. “Telecommunications and Economic Growth: A Panel Data Approach. Applied Economics 36:1649-1654.
- Lam, P.-L., and A. Shiu. 2010. “Economic Growth, Telecommunications Development and Productivity Growth of the Telecommunications Sector: Evidence around the World.” Telecommunications Policy 34(4):185-199.
- Pradhan, R., M. B. Arvin, N. R. Norman, and S. K. Bele. 2014. “Economic Growth and the Development of Telecommunications Infrastructure in the G-20 Countries: A Panel-VAR Approach.” Telecommunications Policy 38(7):634-649.
- Kumar, R. R., R. D. Kumar, and A. Patel. 2015. “Accounting for Telecommunications Contribution to Economic Growth: a Study of Small Pacific Island States.” Telecommunications Policy 39:284-295.
- Norton, S. W. 1992. Transaction Costs, Telecommunications, and the Microeconomics of Macroeconomic Growth.” Economic Development and Cultural Change 41(1):175-196.
- Röller, L.-H., and L. Waverman. 2001. “Telecommunications Infrastructure and Economic Development: A Simultaneous Approach.” American Economic Review 91(4):909-923.
- Lee, S. H., J. Levendis, and L. Gutiérrez. 2012. “Telecommunications and Economic Growth: an Empirical Analysis of Sub-Saharan Africa.” Applied Economics 44(4):461-469.
- Waverman, L., M. Meschi, and M. Fuss. 2005. The impact of telecoms on economic growth in developing countries. Moving the debate forward. The Vodafone Policy Paper Series (2): 10-23.
- World Bank. 2005. World Development Report: A Better Investment Climate for Everyone. World Bank, Washington DC.

### Market Structure, Competition, Network Effects, and Demand Elasticities
- Anderson, S., A. de Palma, and B. Kreider. 2001. “The Efficiency of Indirect Taxes under Imperfect Competition.” Journal of Public Economics 81(2):231-251.
- Auerbach, A., and J. Hines. 2001. “Perfect Taxation with Imperfect Competition.” NBER working paper 8138.
- Averch, H., and L. Johnson. 1962. “The Behavior of the Firm Under Regulatory Constraint.” American Economic Review, December 1962.
- Bertrand, J. 1883. "Théorie mathématique de la richesse sociale.” Journal des Savants, 67: 499–508.
- Corlett, W. J., and D. C. Hague. 1953. “Complementarity and the Excess Burden of Taxation.” Review of Economic Studies 21(1):21-30.
- Davidson, C., and L. Martin. 1985. “General Equilibrium Tax Incidence under Imperfect Competition: A Quantity-Setting Supergame Analysis.” Journal of Political Economy 93(6): 1212-1223.
- Delipalla, S., and M. Keen. 1992. “The Comparison between Ad Valorem and Specific Taxation under Imperfect Competition.” Journal of Public Economics 49:351-367.
- Hamilton, S. 1999. “Tax Incidence under Oligopoly: A Comparison of Policy Approaches.” Journal of Public Economics, 71:233-245.
- Katz, M. L., and H. S. Rosen. 1985. “Tax Analysis in an Oligopoly Model.” Public Finance Quarterly 13(1):3-19.
- Keen, M. 1998. “The Balance between Specific and Ad Valorem Taxation.” Fiscal Studies 19(1): 1-37.
- Economides, N. 1996. “The Economics of Networks.” International Journal of Industrial Organization 14(6):673–699.
- Grajek, M. 2003. “Estimating Network Effects and Compatibility: Evidence from the Polish Mobile Market.” Information Economics and Policy 22:130–143.
- Doganoglu, T., and L. Grzybowski. 2007. “Estimating Network Effects in Mobile Telephony in Germany. Information Economics and Policy 19(1):65-79.
- Garbacz, C., and H. G. Thompson. 2007. “Demand for Telecommunication Services in Developing Countries.” Telecommunications Policy 31:276-289.
- Gasmi, F., M. Ivaldi, and L. R. Virto. 2009. “An Empirical Analysis of Cellular Demand in South Africa.” TSE Working Paper, 09-091.
- Dewenter, R., and J. Haucap. 2008. “Demand Elasticities for Mobile Telecommunications in Austria.” Jahrbücher für Nationalökonomie und Statistik / Journal of Economics and Statistics 228.1:49–63.
- Hakim, S., and S. Neaime. 2014. “The Demand Elasticity of Mobile Telephones in the Middle East and North Africa.” Research in International Business and Finance 32:1-14.
- Karacuka, M., J. Haucap, and U. Heimeshoff. 2011. “Competition in Turkish Mobile Telecommunications Markets: Price Elasticities and Network Substitution.” Telecommunications Policy 35(2):202-210.
- Caves, K. W. 2011. “Quantifying Price-driven Wireless Substitution in Telephony.” Telecommunications Policy 32(11):984-998.
- Sung, N. 2014. “Market concentration and competition in OECD mobile telecommunications markets.” Applied Economics 46(25):3037–3048.
- Symeou, P. 2011. “Economy Size and Performance: An Efficiency Analysis in the Telecommunications Sector.” Telecommunications Policy 35(5):426-440.
- Weyl, G., and M. Fabinger. 2013. “Pass-Through as an Economic Tool: Principles of Incidence under Imperfect Competition.” Journal of Political Economy 121(3): 528-583.
- Hausman, J. 1999. “Efficiency Effects on the U.S. Economy from Wireless Taxation.” Working Paper 7281, NBER Working Paper Series, National Bureau of Economic Research.
- Norton, S. W. 1992. Transaction Costs, Telecommunications, and the Microeconomics of Macroeconomic Growth.” Economic Development and Cultural Change 41(1):175-196.

### Industry Reports, Policy Papers, and Reviews
- Boniecki, D. C. Marcati, W. Abou-Zahr, T. Alatovic and O. El Hamamsy. 2016. “Telecommunications Industry at Cliff’s Edge: Time for Bold Decisions” McKinsey and Company.
- GSMA. 2008. “Comparison of Fixed and Mobile Cost Structures.”
- GSMA. 2012. “The Impact of Taxation on the Development of the Mobile Broadband Sector.”
- GSMA. 2014. “Surtaxes on International Incoming Traffic in Africa.”
- OECD. 2014. International Traffic Termination. OECD Digital Economy Papers No. 238.
- Bauer, J. 2003. “Impact of License Fees on the Prices of Mobile Voice Service.” Telecommunications Policy, 27: 417-434.
- Bazelon, C., and G. McHenry. 2013. “Spectrum Value.” Telecommunications Policy 32:737-747.
- Caves, K. W. 2011. “Quantifying Price-driven Wireless Substitution in Telephony.” Telecommunications Policy 32(11):984-998.
- Koutroumpis, P., A. Lekatsas, G. Giaglis, and P. Kourouthanasis. 2011. “Between a Rock and a Hard Place: Recession and Telecoms Taxation.” Telecommunications Policy 35(7):681-688.
- Prasad, R. 2015. “The Production Function Methodology for Estimating the Value of Spectrum.” Telecommunications Policy 39(1):77-88.
- Schatan, R. 2012. “Tax Minimizing Strategies and the Arm’s-Length Principle.” Tax Notes International, January 9: 121–26.

*Source: wp17247 - REFERENCES (wp17247 - REFERENCES)*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2017/wp17247.pdf_